HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Xu, T.-L. Articles by Akaike, N. Search for Related Content PubMed PubMed Citation Articles by Xu, T.-L. Articles by Akaike, N.

Modulation of the glycine response by Ca²⁺-permeable AMPA receptors in rat spinal neurones

Tian-Le Xu * ¹, Ji-Shuo Li ¹, Young-Ho Jin * and Norio Akaike *

MS 8552 Received 27 July 1998; accepted after revision 22 October 1998.

	ABSTRACT

Top Abstract Introduction Methods Results Discussion References

1. In acutely isolated rat sacral dorsal commisural nucleus (SDCN) neurones, application of kainate (KA) reversibly potentiated glycine-evoked Cl^- currents (I_Gly) in a concentration-dependent manner.

2. The cellular events underlying the interaction between non-NMDA receptors and glycine receptors were studied by using nystatin-perforated patch and cell-attached single-channel recording modes.

3. The action of KA was not accompanied by a shift in the reversal potential for I_Gly. In dose-response curves, KA potentiated I_Gly without significantly changing glycine binding affinity.

4. GYKI 52466 blocked while NS-102 had no effect on the KA-induced potentiation of I_Gly.

5. The potentiation was reduced when KA was applied in a Ca²⁺-free extracellular solution or in the presence of BAPTA AM, and was independent of the activation of voltage-dependent Ca²⁺ channels.

6. Pretreatment with KN-62, a selective Ca²⁺-calmodulin-dependent protein kinase II (CaMKII) inhibitor, abolished the action of KA. Inhibition of calcineurin converted the KA-induced potentiation to a sustained one.

7. Single-channel recordings revealed that KA decreased the mean closing time of glycine-gated single-channel activity, resulting in an increase in the probability of channel opening.

8. It is proposed that Ca²⁺ entry through AMPA receptors modulates the glycine receptor function via coactivation of CaMKII and calcineurin in SDCN neurones. This interaction may provide a new postsynaptic mechanism for control of inhibitory synaptic signalling and represent one of the important regulatory mechanisms of spinal nociception.

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

There is good evidence to suggest that L-glutamate or a related amino acid mediates fast synaptic excitation within the spinal dorsal horn. Glutamate produces its effect through three types of ionotropic glutamate receptors (GluRs), namely -amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA), kainate (KA) and N-methyl-D-aspartate (NMDA). Several classes of GluRs have been immunohistochemically identified in the spinal dorsal horn including AMPA-type GluRs (Yung, 1998). Electrophysiological experiments have shown that fast synaptic excitation of the neurones in the spinal dorsal horn is mediated mainly by activation of AMPA receptors (Yoshimura & Jessell, 1990). These receptors rapidly desensitize when exposed to AMPA, induce non-desensitizing currents when activated by the neurotoxin KA (but see Patneau et al. 1993), and have generally been considered either to be Ca²⁺-impermeable or to have low Ca²⁺ permeability. However, recent studies have demonstrated that several types of neurone, including a subpopulation of neurones in the dorsal horn (Goldstein et al. 1995; Gu et al. 1996; Xu & Akaike, 1996), possess AMPA receptors that are highly Ca²⁺ permeable. Although the functional consequence of Ca²⁺ entry through these receptors is unknown, evidence suggests that, similar to the situation with NMDA receptors, the Ca²⁺ influx may trigger or potentiate intracellular Ca²⁺-dependent processes that could profoundly affect neuronal signal transmission. Thus, activation of Ca²⁺-permeable AMPA receptors results in the strengthening of the synaptic transmission mediated by AMPA receptors (Gu et al. 1996) as well as the inactivation of NMDA receptors in spinal dorsal horn neurones (Xu & Akaike, 1996).

The inhibitory neurotransmission in adult spinal dorsal horn is largely mediated by the strychnine-sensitive glycine (Gly) receptors. The Gly receptors are similar to GABA_A receptors, the major inhibitory amino acid receptors in the central nervous system, in that they both operate a Cl^- channel. Furthermore, previous studies have revealed structural homology between these two receptors and nicotine acetylcholine receptors, and they are considered to be a gene superfamily with a common ancestor (Ortells & Lunt, 1995). There have been a few studies of Ca²⁺-dependent regulation of GABA_A receptors. In bullfrog sensory neurones, for example, the short-term elevation of intracellular Ca²⁺ concentration ([Ca²⁺]_i) produced by activation of voltage-gated Ca²⁺ channels suppresses GABA_A responses (Inoue et al. 1986). The same effect was observed in acutely dissociated hippocampal CA1 pyramidal neurones (Stelzer, 1992; Chen & Wong, 1995) and cultured cerebellar granule cells (Martina et al. 1994). On the other hand, an increase in [Ca²⁺]_i causes a transient augmentation of GABA_A responses in mouse cortical neurones (Aguayo et al. 1998). Recently, the potentiation of GABAergic transmission by presynaptic AMPA receptors was described by Bureau & Mulle (1998). Compared with that of GABA_A receptors, little is known about the intracellular modulation of Gly receptors by Ca²⁺-dependent processes. Some studies suggest an absence of effect of intracellular Ca²⁺ on Gly responses (Tapia et al. 1997), whereas others have reported either an inhibitory (Ragozzino & Eusebi, 1993) or a facilitatory effect of Ca²⁺ on Gly receptors (Kirsch & Betz, 1998).

Previous studies have shown that neurones in the rat sacral dorsal commisural nucleus (SDCN) express Ca²⁺-permeable AMPA receptors as well as strychnine-sensitive Gly receptors (Xu & Akaike, 1996; Xu et al. 1996). This raises the possibility that Ca²⁺-permeable AMPA receptors could serve as a source of Ca²⁺ entry for the induction of Gly receptor regulation in SDCN neurones. Here, the possible interaction between AMPA and Gly receptors in acutely dissociated rat SDCN neurones was studied. Activation of AMPA receptors by KA facilitated the Gly-activated Cl^- currents (I_Gly). This effect was dependent on the Ca²⁺ influx through AMPA receptor channels. Furthermore, the modulation of I_Gly via Ca²⁺ entry through these channels required coactivation of Ca²⁺-calmodulin-dependent protein kinase II (CaMKII) and calcineurin.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Acute isolation of SDCN neurones

The experimental protocol was approved by the Ethics Review Committee for Animal Experimentation of our institute. Wistar rats (2-3 weeks old) were anaesthetized with pentobarbitone sodium (45-50 mg kg^-1, I.P.), and a laminectomy was performed to expose the lower lumbar and sacral spinal cord. A segment about 10-15 mm long of lumbrosacral (L2-S3) cord was dissected out and immersed in freezing incubation solution. The animal was then decapitated. After removal of the attached dorsal rootlets and the pia mater, the spinal segment was affixed with cyanoacrylic glue to a 10 mm × 15 mm agar block which just supported the spinal block. The tissue block together with the agar was positioned in the bottom of the cutting chamber of a vibratome tissue slicer (Dosaka, DTK-100, Kyoto, Japan) with the dorsal face of the cord facing the vibrating blade. A cold incubation solution (5°C; see below) bubbled with 95 % O₂-5 % CO₂ was subsequently poured into the chamber to immerse the spinal block. The spinal segment was sectioned to yield several transverse slices 400 µm thick containing the sacral dorsal commisural nucleus (SDCN) region. The SDCN neurones were acutely dissociated as described elsewhere (Xu et al. 1996). The procedures are essentially identical to the paper by Xu et al. (1996).

Electrophysiology

Electrical measurements were carried out using the nystatin-perforated whole-cell and the cell-attached modes (Hamill et al. 1981) of the patch-clamp technique at room temperature (21-23°C). Patch pipettes were pulled from glass capillaries with an outer diameter of 1·5 mm on a two-stage puller (PB-7, Narishige). The resistance between the recording electrode filled with pipette solution and the reference electrode was 5-7 M. The liquid junction potentials were 3-4 mV, and they were used to calibrate the holding potential (V_h). In single-channel experiments, the pipettes were coated with silicone (Shin-Etsu, Tokyo, Japan) near their tips to reduce their capacitance. The current and voltage were measured with a patch-clamp amplifier (CEZ-2400, Nihon Koden, Tokyo, Japan), filtered at 1 kHz (FV-665, NF Electronic Instruments, Tokyo, Japan), and monitored on both a storage oscilloscope (5100A, Iwatsu Electronic, Tokyo, Japan) and a pen recorder (Recti-Horiz-8K21, Nippondenki San-ei, Tokyo, Japan). Data were simultaneously stored on a videotape after digitization at a rate of 44 kHz (PCM 501 ESN, Nihon Koden). Series resistance, checked every 10 min, was 10-30 M. The change in series resistance during recording was less than 10 %. For single-channel currents, data were filtered using a four-pole low-pass Bessel-type filter (FV-665, NF Electronic Instruments) with a -3 dB corner frequency of 2 kHz, and sampled every 0·1-0·2 ms onto the hard disk of an IBM computer using pCLAMP software (version 6.0, Axon Instruments). The outward currents were shown as upward deflections. In the single-channel analysis, the capacitance current and leak current were removed either by subtracting the averaged current of the blank sweeps or by subtracting a single exponential curve fitted to the closed-channel level. The single-channel amplitudes were measured either by constructing amplitude histograms or by measuring the distance between two horizontal lines set by eye at the open and closed levels. There were no significant differences in the unitary amplitudes determined by these two methods. The channel activities were expressed as the product of the number of functional channels (N) and the open probability (Po). Threshold was set at one-half of the single-channel amplitude, and events above the threshold were regarded as being open. NPo was calculated as the total channel open time divided by the recording time. Ramp voltage commands were generated by a function generator (FG-121B, NF Electronic Instruments). In order to suppress the voltage-dependent Na⁺ and Ca²⁺ currents, 3 × 10^-7 M tetrodotoxin (TTX) and 10^-5 M CdCl₂ were added to the external solution during ramp-wave studies.

Solutions

The composition of the incubation solution was (mM): NaCl, 124; NaHCO₃, 24; KCl, 5; KH₂PO₄, 1·2; CaCl₂, 2·4; MgSO₄, 1·3; and glucose, 10, aerated with 95 % O₂-5 % CO₂ to a final pH of 7·4. The normal external standard solution was (mM): NaCl, 150; KCl, 5; CaCl₂, 2; MgCl₂, 1; Hepes, 10; and glucose, 10. The pH was adjusted to 7·4 with Tris base. The osmolarity of all bath solutions was adjusted to 325-330 mosmol l^-1 with sucrose. The patch-pipette solution for nystatin-perforated patch recording was (mM): caesium methanesulfonate, 100; CsCl, 40; MgCl₂, 5; and Hepes, 10. The pH was adjusted to 7·2 with Tris base. A stock solution of nystatin dissolved in acidified methanol at a concentration of 10 mg ml^-1 was prepared and stored at -20°C. This stock solution was added to the patch-pipette solution to a final nystatin concentration of 200 µg ml^-1 just before use. The ionic composition of the patch-pipette solution for single-channel current recording was (mM): CsCl, 140; tetraethylammonium chloride, 10; CaCl₂, 1; and Hepes, 5, with 3 × 10^-7 M TTX, pH 7·2 with CsOH.

Chemicals

Drugs used in the present experiments were: N-methyl-D-aspartate (NMDA), -amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA), kainate (KA), thermolysin, nystatin, D-2-amino-5-phosphonovalerate (D-AP5), strychnine, forskolin, chlorpromazine, trifluoperazine and W-7 from Sigma; pronase and 1-oleoyl-2-acetylglycerol (OAG) from Calbiochem; W-5 from Molecular Probes; 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) from Tocris Neuramin, Bristol, UK; L-glutamate, cyclothiazide, CYKI 52466, NS-102, chelerythrine and okadaic acid from Research Biochemicals International; N-[2(methylamino)ethyl]-5-isoquinoline sulfonamide dihydrochloride (H-89), KN-62 and KN-04 from Seikagaku corporation, Tokyo, Japan; BAPTA AM from Dojin; and FK506 from Fujisawa Pharmaceutical (Osaka, Japan). Drug solutions were applied by the 'Y-tube' method (Xu et al. 1996) throughout the experiment. This system allows a complete exchange of external solution surrounding a neurone within 20 ms.

Statistical analysis

Experimental values are shown as means ± S.E.M. Statistical significance was determined by one-way analysis of variance (ANOVA) followed by Fisher's post hoc test for multiple comparisons. P and n represent the value of significance and number of experiments, respectively. Values were considered significantly different if P < 0·05.

The Michaelis-Menten equation using a least-squares fitting was applied for evaluation of the half-maximal effective concentration (EC₅₀) of Gly in the concentration-response relationships:

I = Imax C ⁿH/(C ^n_H + Kd^n_H), (1)

where I is the current, Imax is the maximum response, nH is the Hill coefficient, C is the concentration of agonist and Kd is the dissociation constant.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

KA-induced potentiation of IGly

The morphological and electrophysiological features of the isolated SDCN neurones were similar to those reported previously (Xu et al. 1996; Lü et al. 1997). Recordings were made in the perforated-patch configuration which maintains intracellular Ca²⁺ and other second messengers intact (Xu et al. 1996). Exogenous application of 3 × 10^-5 M Gly to a SDCN neurone activated inward Cl^- currents (I_Gly), at a holding potential (V_h) of -50 mV, which were inhibited by the competitive Gly receptor antagonist strychnine in a concentration-dependent manner (Wang et al. 1998). After stable Gly responses were obtained, a solution containing 10^-4 M KA, a non-NMDA receptor agonist, was applied followed by further application of Gly. The amplitude of I_Gly was reversibly potentiated by a preceding KA administration in thirty-two out of forty-eight neurones tested. In another four of these neurones, 10^-4 M KA decreased the peak of I_Gly and this effect was not reversible within the recording period (up to 40 min). KA at the same concentration produced no obvious effect in the other eight neurones. Because of the low incidence of the inhibitory action of KA, we focused only on the KA-mediated facilitation of I_Gly in this study. The potentiation of I_Gly decreased as the interval between KA application and Gly application increased, and the facilitation of I_Gly was no longer detectable when the interval increased to 3 min (Fig. 1A). An interval of 10 s between the application of the two drugs was selected for all the following experiments, since the KA-potentiating effect on I_Gly was sufficiently evident at this time interval. Averaged across all the KA-potentiating neurones, the peak I_Gly was increased to 179 ± 13 % of the control responses (n = 32) under such experimental conditions. The facilitatory effect of KA on I_Gly was concentration dependent with an EC₅₀ of 9·3 × 10^-6 M (Fig. 1B).

		View larger version [in this window] [in a new window]
Figure 1. KA-induced reversible potentiation of IGly Vh = -50 mV. Aa, cumulative records of 3 × 10-5 M Gly-elicited currents (IGly) activated at various times following the conditioning application of 10-4 M KA. Ab, averaged time course of KA-induced facilitation of IGly in 6 cells. IGly was elicited at various intervals indicated by the x-axis following a preceding KA application. The dashed line indicates the control response level induced by 3 × 10-5 M Gly alone. B, concentration-dependent facilitatory effect of KA on IGly. The averaged data were fitted with eqn (1) described in Methods, with an observed EC50 of 9·3 (± 0·7) × 10-6 M (n = 5). Inset, a representative recording corresponding to the concentration-response curve. Here and in subsequent figures the bars above the traces represent the application duration of individual compounds at each concentration indicated (filled bars, Gly; open bars, KA).

Figure 2A shows the current-voltage (I-V) relationship for I_Gly with or without a preceding application of 10^-4 M KA. The reversal potential of I_Gly (V_rev,Gly) was -29·2 ± 1·8 mV (n = 6) in the control and -28·6 ± 2·7 mV (n = 6) with 10^-4 M KA. Both values are close to the Cl^- equilibrium potential (E_Cl) of -29·5 mV calculated by the Nernst equation, suggesting that KA increased the Cl^- currents without affecting other ionic channels. The I_Gly potentiated by KA was completely diminished by adding 10^-6 M strychnine (not shown). In addition, the potentiation induced by KA showed no voltage dependence (Fig. 2A, right-hand lower panel). The potentiating ratio of I_Gly by 10^-4 M KA was independent of Gly concentration. KA increased the maximum value of the concentration- response relationship for Gly without affecting the threshold concentration (Fig. 2B), indicating that KA did not change the apparent affinity of Gly for its receptors. The EC₅₀ value (3·5 × 10^-5 M, n = 6) for Gly in the presence of KA did not significantly differ from 3·7 × 10^-5 M (n = 6) for Gly alone.

		View larger version [in this window] [in a new window]
Figure 2. Effect of KA on the I-V relationship and concentration-response curve for Gly A, I-V relationships for IGly measured in control (a and b) and after KA (10-4 M) conditioning (c and d), by passing a triangular voltage ramp from -40 to -5 mV at a Vh of -50 mV. The figure is representative of 5 reproducible experiments. Vm, membrane potential. B, concentration-response curves for IGly with or without KA (10-4 M) conditioning. Data were obtained from the same neurone. All responses were normalized to the peak current induced by 3 × 10-5 Gly alone (). Similar results were obtained from the other 5 neurones.

Ca²⁺-permeable AMPA receptors are responsible for the potentiation

Both the KA currents and the effect of KA on I_Gly were blocked by the non-NMDA receptor antagonist CNQX. The mean amplitude of the peak I_Gly following KA administration in the presence of 5 × 10^-5 M CNQX was 98 ± 8 % (n = 4) of the control I_Gly. The result demonstrates that this KA effect is mediated through a pharmacologically conventional ionotropic receptor of the non-NMDA subfamily. To establish which subtype of non-NMDA receptor is involved in the modulation of I_Gly, the effects of the AMPA receptor-selective antagonist GYKI 52466 (10^-4 M) and the KA receptor-selective antagonist NS-102 (10^-5 M) were compared for the KA-induced potentiation of I_Gly. The potentiation was completely blocked by GYKI 52466. In addition, application of 10^-4 M AMPA mimicked the KA potentiating effect on I_Gly. This AMPA action was significantly potentiated by cyclothiazide (CTZ), which blocks desensitization at AMPA but not at KA receptors (Fig. 3B). In contrast, NS-102 had no effect on KA-induced potentiation of I_Gly (Fig. 3Aa and B). These data show that the ability of KA to facilitate I_Gly in SDCN neurones is through an action on AMPA rather than KA receptors.

When KA (10^-4 M) was applied in a Ca²⁺-free bath solution, it failed to enhance I_Gly (Fig. 3Ab and B). The facilitatory action of KA on I_Gly also disappeared when the SDCN neurones were loaded with 5 × 10^-5 M BAPTA AM, a membrane-permeable Ca²⁺ chelator, for 2 h (Fig. 3Ac and B). Activation of NMDA receptors results in Ca²⁺ influx, which is known to contribute to many Ca²⁺-dependent processes (Chen & Wong, 1995). Application of 10^-4 M NMDA in the Mg²⁺-free external solution containing 10^-6 M Gly increased I_Gly to a similar extent to 10^-4 M KA application in all seven neurones tested. NMDA (10^-4 M) applied in a Ca²⁺-free bath solution, however, failed to induce the enhancement (Fig. 3B). This gives additional evidence for the possibility of a common facilitatory pathway for the Gly receptors involving Ca²⁺, through the activation of Ca²⁺-permeable AMPA or NMDA receptors. To determine whether the entry of Ca²⁺ exclusively through Ca²⁺-permeable AMPA receptors without activation of voltage-dependent Ca²⁺ channels (VDCCs) was sufficient to induce Ca²⁺-dependent modulation of Gly receptors, we established experimental conditions that precluded the activation of VDCCs by adding 3 × 10^-5 M LaCl₃. Even in the presence of LaCl₃, I_Gly was enhanced by 10^-4 M KA to the same extent as the control condition (Fig. 3Ad and B). In addition, application of 10^-3 M glutamate in the presence of 10^-5 M D-AP5, an antagonist of NMDA receptors, was able to potentiate I_Gly (Fig. 3B). The results indicate that an increase in [Ca²⁺]_i through the activation of AMPA receptors is required for the effect of KA on I_Gly.

		View larger version [in this window] [in a new window]
Figure 3. Pharmacological characterization of the KA-induced potentiation of IGly A, raw current traces showing the antagonism of the effects of 10-4 M KA by Ca2+-free bath solution (b) and 10-5 M BAPTA AM (c), but absence of effect of 10-5 M NS-102 (a) and 3 × 10-5 M LaCl3 (d). B, pooled percentage potentiation of 3 × 10-5 M Gly-evoked currents by 10-4 M KA, 10-4 M AMPA, 10-4 M NMDA or 10-3 M glutamate in the conditions indicated within the corresponding columns. In this and subsequent figures, error bars indicate S.E.M. and the number of the experiments is shown in parentheses. *P < 0·05; **P < 0·01 (ANOVA).

Inhibitors of Ca²⁺-calmodulin and CaMKII prevent the KA-induced potentiation

The present findings suggest that KA potentiated I_Gly in a Ca²⁺-dependent manner. Since previous experiments have shown that intracellular application of the -subunit of CaMKII enhanced I_Gly in spinal dorsal horn neurones (Wang & Randic, 1996), the possible involvement of Ca²⁺- calmodulin (CaM) and CaMKII was examined. CaM inhibitors including chlorpromazine (CPZ), trifluoperazine (TFP) and W-7 were applied for 60 s before the co-administration of 10^-4 M KA and one of the inhibitors. In the presence of these inhibitors, the facilitatory effect of 10^-4 M KA on I_Gly was blocked in a concentration-dependent fashion with IC₅₀ values of 4·3 × 10^-7 M for CPZ, 2·7 × 10^-6 M for TFP and 1·4 × 10^-5 M for W-7. These compounds themselves did not affect I_Gly at all. W-5 used as a negative control of W-7 had no effect (n = 3, not shown). Furthermore, pretreatment for 8 min with 10^-6 M KN-62, a potent inhibitor of CaMKII, markedly abolished the facilitatory effect of 10^-4 M KA on I_Gly. KN-62 itself produced no significant effect on KA-induced currents or I_Gly. Complete recovery from the KN-62 blockade took 10-15 min after washing out the inhibitor (Fig. 4A and B). In contrast, the potentiating effect of KA was not affected by 10^-6 M KN-04, an inactive analogue of KN-62 (Fig. 4B). The results confirm that the Gly receptors in the spinal neurones are modulated by CaMKII and suggest that CaMKII participates in mediating KA-induced enhancement of I_Gly in SDCN neurones.

		View larger version [in this window] [in a new window]
Figure 4. The CaMKII inhibitor KN-62 attenuated KA-induced potentiation A, plot of relative IGly as a function of time and drug application. IGly was potentiated by 10-4 M KA, and 10-6 M KN-62 attenuated the KA action. IGly measured at 20 min after the start of recording () is taken as 1. B, summary of the normalized IGly in the absence () or presence of 10-6 M KN-62 () and 10-4 M KN-04 (). **P < 0·01 (ANOVA).

Our previous investigation demonstrated that I_Gly in the SDCN neurones is regulated by protein kinase C (PKC) and cAMP-dependent protein kinase (PKA) (Xu et al. 1996). It is known that PKC activation could be facilitated by a high [Ca²⁺]_i (Xu et al. 1996). In addition, there is evidence suggesting that CaMKII modulates adenylyl cyclase activity in vivo (Wayman et al. 1995). We therefore performed experiments to examine the possible cross-modulation As summarized in Fig. 5, the facilitatory effect of 10^-4 M KA on I_Gly was not affected by OAG (3 × 10^-6 M), a membrane-permeable PKC activator, chelerythrine (3 × 10^-6 M), a membrane-permeable PKC inhibitor, forskolin (3 × 10^-5 M), a membrane-permeable PKA activator, or H-89 (10^-6 M), a membrane-permeable PKA inhibitor. The data thus suggest that neither PKC nor PKA signalling systems are likely to be involved.

		View larger version [in this window] [in a new window]
Figure 5. No effect of PKC and PKA modulators on the potentiation of IGly by KA A, KA potentiating effect persisted after treatment with 3 × 10-6 M OAG () or 3 × 10-6 M chelerythrine (Chele, ). B, in the presence of 3 × 10-5 M forskolin () or 10-6 M H-89 (), KA (10-4 M) still enhanced IGly. **P < 0·01 (ANOVA).

Addition of okadaic acid and FK506 results in a sustained potentiation of I_Gly

If the potentiation of I_Gly by KA application is due to the activation of CaMKII, as suggested by the above data, then the rapidly reversible nature of the potentiation might reflect transient phosphorylation, owing to endogenous protein phosphatase activity. To test this hypothesis, the effect of okadaic acid (OA), a non-selective protein phosphatase inhibitor, was examined. Loading the neurones with a concentration of OA (10^-6 M) that inhibits protein phosphatase (PP)1 and PP2A but not PP2B (calcineurin) (Bialojan & Takai, 1988) had no significant effect on the baseline I_Gly and the KA-induced potentiation of I_Gly. In the presence of 10^-6 M OA, pre-application of 10^-4 M KA still reversibly increased I_Gly to 173 ± 7 % of the control responses (n = 4). A higher concentration of OA (5 × 10^-6 M) that inhibits calcineurin, however, converted the KA-induced potentiation to a sustained one as well as enhanced the baseline I_Gly to 136 ± 14 % of the control I_Gly (Fig. 6; n = 5, P < 0·05). During the potentiation of I_Gly by OA, application of 10^-4 M KA resulted in an irreversible and larger enhancement of I_Gly (Fig. 6Ab and B). OA at the concentrations used produced no significant effect on the KA-induced currents (Fig. 6Aa).

To confirm that calcineurin was the protein phosphatase inhibited in the above experiment, the effect of FK506, a potent and selective calcineurin inhibitor, was examined. Application of 10^-6 M FK506 prolonged the KA-induced potentiation of I_Gly and enhanced the baseline I_Gly (n = 7; Fig. 6B), the observed effects being comparable to those seen in the presence of 5 × 10^-6 M OA (Fig. 6B). The results thus suggest that endogenous calcineurin activity can modulate I_Gly and that calcineurin might be involved in the mechanism for rapid recovery from the KA potentiating effect.

		View larger version [in this window] [in a new window]
Figure 6. KA-induced potentiation was not transient in the presence of calcineurin inhibitors Aa, raw current traces at the times indicated in Ab. Ab, plot of relative IGly as a function of time and drug application. The IGly measured at 20 min after start of recording () is taken as 1. B, summary of the effects of calcineurin inhibitors. The dashed line indicates the control response level induced by 3 × 10-5 M Gly alone. *P < 0·05; **P < 0·01; ***P < 0·001 (ANOVA).

Modulation of Gly single-channel activities by KA in the cell-attached configuration

In the cell-attached mode, single-channel activities were constantly recorded when the patch pipette contained 10^-6 M Gly (15 out of 15 patches). There was no channel activity when the patch pipette contained no Gly (8 out of 8 patches). The main level, which occurred more than 95 % of the time, had a mean size of 1·45 ± 0·11 pA (n = 8) at a pipette potential (Vp) of 0 mV. The sublevel had a mean amplitude of 1·00 ± 0·01 pA (n = 8). With the resting potential around -51 mV, the estimated conductances were 28·4 and 19·6 pS, respectively, which corresponded well with the conductances of single Gly-activated channels in spinal neurones (Bormann et al. 1987). KA (10^-6 M) had no effect on the Gly single-channel conductances in all eight neurones tested, whereas the probability of channel opening (Po) was increased in five out of eight experiments (Fig. 7 and Table 1). In the present experimental conditions, no channel activities were observed after application of 10^-6 M KA when the patch pipette contained no Gly (10 out of 10 patches). The relatively low concentration of KA was chosen so as to limit soma depolarization and changes in the resting potentials. Table 1 summarizes the effect of 10^-6 M KA on the main kinetic parameters of Gly receptor channels. KA decreased the mean closing time of Gly-gated single-channel activity, thus resulting in an increase in the probability of channel opening.

		View larger version [in this window] [in a new window]
Figure 7. Cell-attached recordings of Gly-activated single-channel activities at a pipette potential (Vp) of 0 mV Recording pipette contained 10-6 M Gly. Outward currents appear as upward deflections. The resting membrane potential was -51 mV. In the control solution, the main patch current amplitude was 1·47 ± 0·13 pA, and probability of opening (Po) was 0·001 ± 0·004. After adding 10-6 M KA, the current was 1·45 ± 0·14 pA, and Po = 0·004 ± 0·009. A, slow time base. B, fast time base.

Table 1. Glycine-induced single-channel currents